The Design And Implementation Of The Arrest And Distribution Control System Based On Big Data Analysis

At present,Mianyang City has built more than 700 highway checkpoints,recording more than 6 million data and information every day,accompanied by a large number of pictures and video data.However,as more and more newly built checkpoint systems are connected to the original arrest and distribution control system,the problems of the system become more and more serious,so the original arrest and distribution control system needs to be improved.In the original system,the servlet interface is used to receive the data uploaded from the checkpoint system,and then the data is stored in the Oracle database after comparison and analysis.This architecture leads to a very serious problem.Because the data in the interface will be compared and analyzed,there is a difference between the speed of data processing in the interface and the speed of data uploading in the checkpoint system,so it is easy to cause data backlog in the interface,resulting in very low real-time data at the business end.When the number of checkpoints in the original system is small,this problem is not very obvious.But with the development of the city,the number of checkpoints is growing,and the problem is becoming more and more serious.At the beginning of the problem,the solution provided was to increase the number of servers in the system,and the problem has been solved temporarily.However,with the increase of data volume in the relational database,the query and statistical efficiency of the business end has also been greatly reduced,which can not meet the normal business requirements of the business system.Therefore,through the analysis,the paper puts forward the Hadoop framework and big data analysis method to build the arrest and distribution control system.In the face of a large number of passing record data,the system uses big data storage and big data analysis technology to process it.The system caches the passing data uploaded by the checkpoint system through Kafka distributed message queue,and returns the data to upload successfully without any processing,which solves the problem of passing data accumulation at the interface.Through the spark streaming flow calculation,the system can quickly analyze the passing data and improve the real-time performance of the system.The system stores the passing data through the HBase distributed database,so that the business function can reach the second level response.The system adopts the AngularJS architecture at the web end and the Spring MVC framework at the back end,making full use of the storage and computing capacity of big data,greatly improving the police user experience.Compared with the original arrest and distribution control system,the system uses big data analysis to improve the access and analysis efficiency of passing data,optimize the system structure,improve the user experience,and ensure the practicability of the system.At present,the research results of this paper have been applied in the road intelligent transportation big data center of Mianyang traffic police detachment,and the relevant functions of the vehicle for arrest,distribution and control and statistical analysis have been realized,which is of great significance to promote the construction of "China Science and technology city" and "smart city" in Mianyang.